EP2406903A1 - Signatures numériques - Google Patents

Signatures numériques

Info

Publication number
EP2406903A1
EP2406903A1 EP10750290A EP10750290A EP2406903A1 EP 2406903 A1 EP2406903 A1 EP 2406903A1 EP 10750290 A EP10750290 A EP 10750290A EP 10750290 A EP10750290 A EP 10750290A EP 2406903 A1 EP2406903 A1 EP 2406903A1
Authority
EP
European Patent Office
Prior art keywords
data segment
signature
value
threshold value
operable
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Withdrawn
Application number
EP10750290A
Other languages
German (de)
English (en)
Other versions
EP2406903A4 (fr
Inventor
Ravosh Samari
Paymaan Behrouzi
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Individual
Original Assignee
Individual
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Individual filed Critical Individual
Publication of EP2406903A1 publication Critical patent/EP2406903A1/fr
Publication of EP2406903A4 publication Critical patent/EP2406903A4/fr
Withdrawn legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L9/00—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols
    • H04L9/32—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols including means for verifying the identity or authority of a user of the system or for message authentication, e.g. authorization, entity authentication, data integrity or data verification, non-repudiation, key authentication or verification of credentials
    • H04L9/3247—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols including means for verifying the identity or authority of a user of the system or for message authentication, e.g. authorization, entity authentication, data integrity or data verification, non-repudiation, key authentication or verification of credentials involving digital signatures
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04H—BROADCAST COMMUNICATION
    • H04H60/00—Arrangements for broadcast applications with a direct linking to broadcast information or broadcast space-time; Broadcast-related systems
    • H04H60/35—Arrangements for identifying or recognising characteristics with a direct linkage to broadcast information or to broadcast space-time, e.g. for identifying broadcast stations or for identifying users
    • H04H60/37—Arrangements for identifying or recognising characteristics with a direct linkage to broadcast information or to broadcast space-time, e.g. for identifying broadcast stations or for identifying users for identifying segments of broadcast information, e.g. scenes or extracting programme ID
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04H—BROADCAST COMMUNICATION
    • H04H60/00—Arrangements for broadcast applications with a direct linking to broadcast information or broadcast space-time; Broadcast-related systems
    • H04H60/56—Arrangements characterised by components specially adapted for monitoring, identification or recognition covered by groups H04H60/29-H04H60/54
    • H04H60/58—Arrangements characterised by components specially adapted for monitoring, identification or recognition covered by groups H04H60/29-H04H60/54 of audio
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L9/00—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols
    • H04L9/32—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols including means for verifying the identity or authority of a user of the system or for message authentication, e.g. authorization, entity authentication, data integrity or data verification, non-repudiation, key authentication or verification of credentials
    • H04L9/3247—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols including means for verifying the identity or authority of a user of the system or for message authentication, e.g. authorization, entity authentication, data integrity or data verification, non-repudiation, key authentication or verification of credentials involving digital signatures
    • H04L9/3255—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols including means for verifying the identity or authority of a user of the system or for message authentication, e.g. authorization, entity authentication, data integrity or data verification, non-repudiation, key authentication or verification of credentials involving digital signatures using group based signatures, e.g. ring or threshold signatures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/60—Information retrieval; Database structures therefor; File system structures therefor of audio data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/60—Information retrieval; Database structures therefor; File system structures therefor of audio data
    • G06F16/68—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/683—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
    • G06F16/78—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/783—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04H—BROADCAST COMMUNICATION
    • H04H2201/00—Aspects of broadcast communication
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04H—BROADCAST COMMUNICATION
    • H04H2201/00—Aspects of broadcast communication
    • H04H2201/90—Aspects of broadcast communication characterised by the use of signatures
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L9/00—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols
    • H04L9/32—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols including means for verifying the identity or authority of a user of the system or for message authentication, e.g. authorization, entity authentication, data integrity or data verification, non-repudiation, key authentication or verification of credentials
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L9/00—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols
    • H04L9/32—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols including means for verifying the identity or authority of a user of the system or for message authentication, e.g. authorization, entity authentication, data integrity or data verification, non-repudiation, key authentication or verification of credentials
    • H04L9/3247—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols including means for verifying the identity or authority of a user of the system or for message authentication, e.g. authorization, entity authentication, data integrity or data verification, non-repudiation, key authentication or verification of credentials involving digital signatures
    • H04L9/3257—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols including means for verifying the identity or authority of a user of the system or for message authentication, e.g. authorization, entity authentication, data integrity or data verification, non-repudiation, key authentication or verification of credentials involving digital signatures using blind signatures

Definitions

  • the present invention relates to digital signatures, which are also called “digital fingerprints” herein, for example digital audio signatures (digital audio fingerprints), and to apparatus and methods for generating digital signatures (digital fingerprints).
  • audience measurement/analysis is audience measurement/analysis, as explained below.
  • Audience measurement/analysis systems for measuring/analysing TV or radio audiences for instance, commonly make use of so-called meters installed in a panel of households, or on a panel of persons, generally chosen such that the household occupants or persons are demographically representative of the potential audience population as a whole.
  • the meters are devices which monitor the channels, stations or programs selected for viewing or listening on a TV or radio in the household, or by the person, and typically record information concerning the selected channels, stations or programs for sending, for example by telephone line or other means of communication, to a "central" or “reference” office at which viewing/listening information from households in the panel is collected for analysis.
  • references information may also be obtained from other sources. For example information concerning programs broadcast on a particular channel or station over a period of time may be obtained directly from the broadcasting company.
  • the information from households/persons may then be analysed to reveal channels, stations or programs selected for viewing in the households or by persons, by comparison or matching of information from the households/persons with the reference information.
  • One technique used at least in households is to directly monitor tuning circuits in a
  • Another technique is to monitor special identification codes embedded in program signals, for example as broadcast or delivered by a service provider.
  • the codes are embedded "at source" in the program signal by the service provider, for example in an audio signal component, or in an video signal component if video is involved.
  • Such embedded codes typically identify the broadcasting station, and may also containing information identifying the program carried by the station at any given time.
  • a third technique is to derive, from a program selected for viewing or listening, a signature (i.e. fingerprint) characteristic of the program.
  • the signature may be derived from audio or video.
  • Such a signature is not embedded in the program signal at source, but is derived or generated from the program signal at the point of viewing or listening in a household or by a person.
  • signatures/fingerprints has been considered to be demanding in terms of cost and the hardware needed, and susceptible in practice to program identification problems due to inconsistencies between signatures/fingerprints generated, on the basis of the same original program content, under the different conditions which may apply at different locations and different times.
  • the different conditions may arise as a result of different program reception conditions or equipment capabilities at different times or locations, for example in a household or at a person on the one hand and at a central or reference office on the other hand.
  • a signature/fingerprint or a sequence of signatures/fingerprints, for example representing a section of audio, from within reference information or a reference database, e.g. of audio, available at a central or reference office.
  • a reference database e.g. of audio
  • the fingerprints should allow searching from one set of fingerprints (e.g. derived in a household or at a person) within another set (e.g. available at a central or reference office), with the goal of determining if fingerprints from one set match any fingerprints from the other set thereby determining a match.
  • one set of fingerprints e.g. derived in a household or at a person
  • another set e.g. available at a central or reference office
  • a user or audience member is exposed to audio, for example, produced from television, radio, internet, or similar broadcast.
  • DRM Digital Rights Management
  • Yet another field of the invention is in royalty distribution.
  • media broadcasts utilize copyright material.
  • the artist or publisher of this material allows their art to be broadcast usually in return for compensation.
  • royalty distribution organizations that collect funds from the broadcasters or users of the material and distribute the funds to the artists and/or publishers of the material depending on the usage.
  • Digital signatures/fingerprints can be used to identify which copyright material is being broadcast such that the artist or publisher of the material will be duly compensated.
  • the present invention is concerned with the provision of robust signatures/fingerprints. Aspects of the invention are indicated in the accompanying claims and in the paragraphs at the end of this description.
  • a series of digital fingerprints can be generated, both from original broadcast content, e.g. audio content, and secondarily, from the audio produced by audience receiving equipment.
  • original broadcast content e.g. audio content
  • secondarily from the audio produced by audience receiving equipment.
  • a match comprising either an exact match or a close match of the fingerprint, will indicate which media the user was exposed to and therefore also which broadcast.
  • the apparatus for sampling the audio that the user (audience member) is being exposed to can either be electrically coupled signal pickup, or it can be via a microphone device that will record the audio signals in the environment.
  • the invention may be used to generate digital fingerprints from all available copyright material, and likewise the same methods and apparatus may be used to generate digital fingerprints from material posted on the internet for example. The fingerprints are compared and if a match is found, the media content posted on the internet will be identified and will allow the owner of the copyright material to police its usage.
  • An original signal e.g. audio, and broadcast or received signal may be significantly different due to various distortions including in the case of audio ambient noise, reverb, compression/decompression distortion, playback speed variations, tone adjustment distortion, etc.
  • Method and apparatus in accordance with the invention used to digitally fingerprint signals, e.g.
  • the present invention can provide for satisfactory match rates. Further, the present invention can avoid the disadvantage of some current systems is that a large size of fingerprint is needed and the volume of fingerprints makes data storage, movement, and searching problematic.
  • embodiments of the present invention can provide for the efficient generation of essentially unique digital fingerprints from segments of audio. These fingerprints can then be used to identify the audio even if the original audio is different from the audio to be matched due to various distortions.
  • FIGURE 1 is a schematic illustration of apparatus embodying the present invention used to capture audio, then generate and store digital fingerprints from captured audio samples.
  • FIGURE 2 is a schematic flow diagram of steps used in generating digital fingerprints in accordance with an embodiment of the present invention.
  • FIGURE 2a is a schematic flow diagram of steps used in generating digital fingerprints in accordance with another embodiment of the present invention.
  • FIGURE 3 is a diagram illustrating the operation of a WAVEFORM SHAPER
  • FIGURE 4 is a diagram illustrating the operation of a THRESHOLD DETERMINATION function in accordance with an embodiment of the present invention.
  • FIGURE 4a is a diagram illustrating the operation of a THRESHOLD DETERMINATION function in accordance with another embodiment of the present invention.
  • FIGURE 5 is a diagram illustrating the operation of a FINGEPRTNT START POINT IDENTIFIER function in accordance with an embodiment of the present invention, which selects starting points for fingerprint generation.
  • FIGURE 6 is a diagram illustrating a SPECTRAL FINGERPRINT GENERATOR operation in accordance with an embodiment of the present invention.
  • FIGURE 7 is a diagram illustrating a TIME-DOMAIN FINGERPRINT GENERATOR operation in accordance with an embodiment of the present invention.
  • FIGURE 7a is a diagram illustrating a TIME-DOMAIN FINGERPRINT GENERATOR operation in accordance with an embodiment of the present invention.
  • FIGURE 8 is a diagram showing a JOINER operation in accordance with an embodiment of the present invention, which uses both the spectral fingerprint and time domain fingerprint to produce several new fingerprints.
  • the audio capture device used may be a sound card in a computer, or other audio digitizing equipment.
  • the audio capture device may capture sound waves via a transducer, such as a microphone, and digitize via known methods.
  • apparatus as illustrated may for example by used in an audience measurement system to capture broadcast audio that a user is exposed to such as from a TV set, a radio, or a computer viewing media either from the internet, or locally stored and also to capture audio used to generate a reference database.
  • Reference audio used to generate the reference database, can originate for example from
  • FIGURE 1 shows apparatus for use in capturing and generating fingerprints.
  • the audio capture device 100 may, as indicated above, either capture the audio electrically via electrical coupling, or capture via sound waves, via transducer or microphone, or the audio can be presented as an already digitized audio file.
  • the sampling frequency or rate used which preferably should stay fixed across the system, can be selected as appropriate.
  • a 8000Hz sampling rate is used, for example with 16 bit resolution per sample.
  • the digitized audio samples are filtered using a band pass filter arrangement 200.
  • a separate and different band pass filter 1000 is used to filter audio for optional spectral fingerprint generation, which will be explained below.
  • the band pass filter arrangement 200 may comprise a number, for example four, band pass filters having different pass bands. It may be provided that the pass bands of the filters can be adjusted, to provide for optimal performance. Examples of pass bands for filtering audio samples are: 300 to 800Hz; 800 to 1300Hz; 1300 to 1800Hz; 1800 to
  • a waveform shaper 300 is employed to shape the waveform from the filter in a two step process which is illustrated in FIGURE 3.
  • FIGURE 3 shows three stages, A, B, and C, in the waveform shaping.
  • Stage A shows the original (filtered) digitized audio.
  • Stage B shows the result after the first 'shaper' step has been performed.
  • Stage C shows the final result of the WAVEFORM SHAPER function.
  • the first step involves removing negative content from the original digitized audio waveform. This is achieved in two possible ways. The first option is by replacing with 0 all values below zero. The second option, used in a preferred embodiment of the invention, is to convert negative samples to their positive equivalent absolute values. The resulting waveform, showing negative content removed, is shown in stage B of FIGURE 3.
  • the second step divides the result of stage B into time blocks consisting of a predetermined number of samples (referred to as blocks of W size in step 2300 of FIGURE 2). In a preferred embodiment this predetermined number of samples is 500. Successive values of the samples in each block are then summed up and the result is then used as the value for the entire block. Each block is summed separately.
  • the resultant waveform is shown as stage C in FIGURE 3, and referred to below as the Pseudo Envelope Waveform or Pseudo Envelope Array (PE). If the output of the bandpass filter arrangement 200 is a plurality of waveforms, a plurality of PE's are provided.
  • FIGURE 2 This steps or functions effected so far are shown in FIGURE 2 at 2000, 2100, 2200, 2300, 2400, and 2500.
  • a PE waveform or PE array is then passed to a THRESHOLD DETERMINATION block 400 as illustrated in FIGURE 1.
  • FIGURE 4 illustrates the detailed operation of the THRESHOLD DETERMINATION block for an embodiment of the invention. The operation is performed by dividing the PE into blocks of time.
  • the block size (referred to as TW in step 2600 of FIGURE 2) is greater than the block size (W) used in the PE (WAVEFORM SHAPER) section.
  • the block size is chosen to be a few seconds of audio. In an implementation with a sample rate of 8000Hz, the block size in a preferred implementation is 60,000 samples.
  • a par value is calculated.
  • This par value may be for example an average, mean, middle or median value of the samples in the block.
  • An easy way to determine a median value is to sort all values and select the middle value, however any method to determine the median or par value is acceptable.
  • MF a factor
  • this factor is in the range of 1.00 to 1.30. This factor is of significance for the Fingerprint Start Point Identifier function which is described in more detail below.
  • THRESHOLD DETERMINATION is also illustrated in FIGURE 2 at 2600 and 2700.
  • each sample in a PE waveform or array will also have an associated threshold value (par value multiplied by the factor). It is noted that many samples in a PE can share the same threshold value.
  • the threshold values may be provided by using a form of moving threshold.
  • the threshold for a sample may be the threshold determined for the temporally preceding sample or block multiplied by a multiplication factor less than unity, for example 0.8, the product of which multiplication is added to the product of the current sample value, or the current block par value, multiplied by unity minus the multiplication factor.
  • threshold value determination is illustrated in the flowchart of FIGURE 2a (see in particular 2600/2700) and in FIGURE 4a, where the resulting moving threshold is illustrated.
  • each is subject to threshold determination as described above. For example the same or different block sizes, multiplication factors etc. may be applied to threshold determination for different PE's.
  • FIGURE 5 illustrates detailed operation of the FINGERPRINT START POINT IDENTIFIER 500 shown in FIGURE 1.
  • the FINGERPRINT START POINT IDENTIFIER identifies starting points where fingerprints (fingerprint generation) will be initiated. The operation is indicated in FIGURES 2 and 2a, 2800.
  • a fingerprint start point also called a Sync Event or simply an Event, is determined at each point where the value in the PE array transitions from below the corresponding threshold value to above the threshold - as indicated with arrows in FIGURE 5.
  • the FINGERPRINT START POINT IDENTIFIER may also be called an event detector.
  • the operation of establishing a threshold value depends on a multiplication factor. Increasing the value of this multiplication factor can be expected to decrease the number of the fingerprint start points or events, whereas decreasing the multiplication factor can be expected increase the number of fingerprint start points or events. Increasing the number of fingerprint start points or events will increase the robustness and noise immunity of the system.
  • threshold determination as described in connection with FIGURES 2 and 4, or threshold determination as described in connection with FIGURES 2a and 4a it is possible to reduce the number of fingerprint start points or events by selecting only those start points or events which are closest to (closest before or closest after) a major peak in the PE waveform. For example a limited number of only the highest peaks in the waveform may be considered.
  • each is subject to determination of fingerprint starting points (Sync Events) as described above.
  • Fingerprint starting points determined as described above are used in relation to generation of both time-domain fingerprints (see steps 2900 and 3800 to 4300 in FIGURES 2 and 2a) and spectral digital fingerprints (See steps 3000 to 3700 in FIGURES 2 and 2a).
  • Generation of spectral fingerprints may be based on the original signal (unfiltered) digitized signal or the original signal after passage through a band pass filter arrangement 1000, shown in Figure 1 (Band Pass Filter 2 in 3200 of FIGURES 2 and 2a), rather than the filter arrangement 200 (Band Pass Filter 1 in 2100 of FIGURES 2 and 2a), as explained below.
  • the optional spectral fingerprint may be generated using frequency analysis.
  • this spectral fingerprint is generated by using the same fingerprint start points, performing a Fast Fourier Transform, or other technique to determine spectrum power on the original, unfiltered input signal (e.g. audio).
  • 4000 samples of audio were used for each FFT calculation, which is equivalent to half a second of audio.
  • spectrum information below 800Hz was considered. This was considered to be advantageous because it reduces greatly the processing of the FFT.
  • the chosen spectrum is divided into sections. In the example implementation 5 sections are used. In each section, the frequency with the highest power was identified and a 5 bit digital pattern attributed to it.
  • the resulting 5 sequences of 5 bits each are concatenated together to generate a 25 bit fingerprint.
  • the digital pattern is chosen such that neighbouring items will only differ by one bit.
  • the use of spectral signatures is optional. Experimentation has shown that acceptable results can be provided using only time domain fingerprints, without using spectral fingerprints. However in some situations use of spectral fingerprints may enable improved overall fingerprint matching rate.
  • FIGURE 6 illustrates detailed operation of the SPECTRAL FINGERPRINT GENERATION 600 shown in FIGURE 1 , in accordance with another embodiment of the invention.
  • the starting point for each fingerprint is determined in the FINGERPRINT START POINT LOCATION IDENTIFIER block, as explained above.
  • a Spectral Fingerprint is generated from the raw digitized audio after it has been filtered using the bandpass filter arrangement 1000 as shown in FIGURE 1 and indicated in FIGURES 2 and 2a at steps 3000, 3100 and 3200.
  • the audio sample size consists of 4000 samples equating to a half a second of audio.
  • the frequency content of this snippet of audio is calculated for example using Fast Fourier Transform, or Discrete Fourier Transform, or any other method that indicates the energy in the spectrum (step 3300 in FIGURES 2 and 2a).
  • the spectrum is divided into blocks. It is possible to use 1, 2, 3, or 4, or more blocks, but in a preferred form of this embodiment of this invention, 4 blocks are used (step 3400 in FIGURES 2 and 2a).
  • each block contains a number of frequency bins or divisions (step 3500 in FIGURES 2 and 2a). In a preferred form of this embodiment, 255 bins are used for each block.
  • the bin or division with the maximum energy from each block is taken (step 3600 in FIGURES 2 and 2a).
  • each block will produce a number between 1 and 255 corresponding to the bin with the maximum energy.
  • Each number is then converted to its binary equivalent and the binary numbers are concatenated (step 3700 in FIGURES 2 and 2a).
  • the final result is a binary fingerprint consisting of many bits.
  • SF spectral fingerprint
  • spectral fingerprints can be derived for a plurality of frequency bands, for instance 300Hz to 800Hz, from 800Hz to 1300Hz etc..
  • FIGURE 7 shows the detailed operation of the TIME-DOMAIN FINGERPRINT
  • TIME-DOMAIN FINGERPRINT TDF
  • Tstep the value in the PE array, at a location that is a forward offset according to previously chosen offset size.
  • the offset can be any value, however, in a preferred embodiment, it (Tstep) is chosen to be 3000 samples. The value in the PE at this offset is compared to the value of the threshold for that location.
  • next bit in the fingerprint should be a 1 O'. Otherwise, it is understood that the next bit is a '1' (step 4200 in FIGURES 2 and 2a). As depicted in FIGURE 7, this process continues until all the required bits are collected and a full fingerprint is thus generated (step 4300 in FIGURES 2 and 2a).
  • the number of bits in the fingerprint can be any previously chosen value, however, in preferred implementations, 30 or 40 bits have been chosen.
  • FIGURE 7a illustrates the obtaining of time domain fingerprints using a moving threshold as illustrated in FIGURE 4a (and FIGURE 2a).
  • using an individual identified starting point identified in the previous stage 40 bits are collected.
  • the 40 bits are collected by jumping out a predetermined number of samples for each bit to be collected and comparing the value of the PE at that location to the value moving threshold at that location. If the value PE is higher then the corresponding bit in the fingerprint is "1", otherwise it is "0".
  • the generated fingerprints i.e. time domain fingerprints and, if generated, spectral fingerprints may then be sent for storage in fingerprint storage 900 as shown in Figure 1 (e.g. step 4600 in FIGURES 2 and 2a).
  • fingerprints may be joined or concatenated as explained below. Such joining or concatenation may not be used in some embodiments of the invention, but in other embodiments may be of advantage depending on the application concerned.
  • FIGURE 8 shows the detailed operation of the JOINER 800 illustrated in FIGURE 1.
  • a SPECTRAL FINGERPRINT (SF) and a TIME-DOMAIN FINGERPRINT (TDF) may be joined together or concatenated (step 4400 in FIGURES 2 and 2a) in a variety of ways to ultimately produce several fingerprints all of which represent the same original signal (e.g. piece of audio).
  • the fingerprints to be joined or concatenated in each case relate to the same fingerprint starting point (FSP), and concatenation may be carried out in respect of all FSP's (step 4500 in FIGURES 2 and 2a).
  • FSP fingerprint starting point
  • FIGURE 8A one fingerprint is generated by only taking a portion of an SF and then concatenating that with an entire TDF.
  • FIGURE 8B another fingerprint is generated by taking the entire SF and concatenating that with a portion of the TDF.
  • yet another fingerprint is generated by taking a portion of the SF and concatenating that with a portion of the TDF.
  • portions or all of each type of fingerprint can be used in the preparation of final fingerprints. This may provide for advantage in the robustness of the fingerprints such that ultimately the fingerprints will be more immune to noise and distortion.
  • one of the fingerprints might work ideally with signals subject to one type of distortion
  • another type of fingerprint may work ideally with another type of distortion.
  • time domain fingerprints generated from the four PE waveforms and the spectral fingerprint or fingerprints from the FFT stage - associated with each individual fingerprint starting point - may be concatenated to form one large digital signature.
  • a final concatenated fingerprint consisted of 40 bits for each time domain fingerprint and 25 bits for the (single) spectral fingerprint. This adds up to 185 bits. Testing has shown that signatures as small as 70 bits and as large as 500 bits are feasible. However a large size of fingerprint may be considered a disadvantage in some applications, the volume of fingerprints possibly making data storage, movement, and searching problematic.
  • digital signals regardless of their information content, can be subject to fingerprint or signature generation in accordance with the present invention.
  • digital signals representing any form of scanned image may be subject to fingerprint generation.
  • the image itself might be an image of a face or an image of a true fingerprint or handprint, or an image of a banknote, for example for identification or verification purposes.
  • the present invention comprehends the aspects set out in the following paragraphs :-
  • Digital signature generation apparatus comprising an envelope generator operable to generate an envelope representation, of only one polarity, of a sampled data segment, the envelope generator comprising (i) a polarizer operable (a) to remove samples in the data segment which are of polarity opposite to the one polarity, or (b) to reverse the polarity of samples in the data segment which are of polarity opposite to the one polarity, to provide a polarized data segment and
  • a sample value summer operable, for each of successive portions of the polarized data segment, each portion comprising a predetermined plurality of samples, to provide a portion sum value as the sum the values of the samples in the portion, thereby to provide said envelope representation;
  • a threshold value generator operable to determine threshold values for each of successive blocks of the envelope representation, each block comprising a plurality of the successive portions, each threshold value being determined based on a par value of the portion sum values of the portions comprised in the block concerned; an event detector operable to detect, as an event, a transition of a portion sum value across the threshold value for the block to which the portion concerned belongs; and a signature generator operable in response to the or a detected event to a generate a digital signature characteristic of the sampled data segment.
  • Digital signature generation apparatus comprising an envelope generator operable to generate an envelope representation, of only one polarity, of a sampled data segment, the envelope generator comprising
  • a polarizer operable (a) to remove samples in the data segment which are of polarity opposite to the one polarity, or (b) to reverse the polarity of samples in the data segment which are of polarity opposite to the one polarity, to provide a polarized data segment and
  • a sample value summer operable, for each of successive portions of the polarized data segment, each portion comprising a predetermined plurality of samples, to provide a portion sum value as the sum the values of the samples in the portion, thereby to provide said envelope representation;
  • a threshold value generator operable to determine a threshold value for each portion of the envelope representation, each threshold value being derived from (A) the threshold of the preceding portion, and (B) the the current portion sum value and (C) a determined multiplication factor;
  • an event detector operable to detect, as an event, a transition of a portion sum value across the threshold value for the portion concerned; and a signature generator operable in response to the or a detected event to a generate a digital signature characteristic of the sampled data segment.
  • each threshold value is the sum of (i) the threshold of the preceding portion multiplied by a multiplication factor less than unity, and (ii) the product of the current portion sum value and unity minus said multiplication factor.
  • AE Apparatus in accordance with any of paragraphs (AA), (AB), (AC) and (AD), wherein the signature generator is operable, on the basis of a predetermined signature collection pattern specifying a plurality of sample offsets from the or a detected event, to determine whether the portion sum value at each sample offset is greater or less than the threshold value at the sample offset concerned, and to set the value of a corresponding bit of a signature in dependence thereupon, to provide a binary signature.
  • AH Apparatus in accordance with any of paragraphs (AA), (AB), (AC), (AD), (AE), (AF) and (AG), wherein the signature generator is operable, in response to the or a detected event, to determine energy content across a frequency spectrum of a section of the sampled data segment, to divide the frequency spectrum into a plurality of frequency bands which are allocated respective identifying numbers, and to determine the identifying number of the frequency band of maximum energy, to provide a frequency signature comprising this identifying number.
  • AI Apparatus in accordance with any of paragraphs (AA), (AB), (AC), (AD), (AE), (AF) and (AG), wherein the signature generator is operable, in response to the or a detected event, to determine energy content across a frequency spectrum of a section of the sampled data segment, to divide the frequency spectrum into a plurality of frequency bands which are allocated respective identifying numbers, and to determine the identifying number of the frequency band of maximum energy, to provide a frequency signature comprising this identifying number.
  • AI Appa
  • AK Apparatus in accordance with any of paragraphs (AH), (AI) and (AJ) which is also in accordance with any of paragraphs (AE), (AF) and (AG), wherein the signature generator is operable to provide a signature comprising in combination at least parts of the binary signature and the frequency signature.
  • the threshold value generator further comprises a multiplier operable to multiply the par value by a predetermined multiplication factor, to provide the block threshold value.
  • A Apparatus in accordance with any of paragraphs (AA), (AB), (AC), (AD), (AE), (AF), (AG), (AH), (AI), (AJ), (AK) and (AL), further comprising a filter operable to band-pass filter the sampled data segment, the envelope detector generating the envelope representation on the basis of the band-pass filtered data segment.
  • AN Apparatus in accordance with paragraph (AM), wherein the filter has a plurality of pass bands and the envelope detector is operable to generate respective envelope representations on the basis of the data segment as filtered in each pass band.
  • AO Apparatus in accordance with any of paragraphs (AA), (AB), (AC), (AD), (AE), (AF), (AG), (AH), (AI), (AJ), (AK), (AL), (AM) and (AN), wherein the event detector is operable to detect, as an event, for each envelope representation generated on the basis of the data segment as filtered in each pass band, a transition of a portion sum value across the threshold value for the portion concerned.
  • AQ Apparatus in accordance with any of paragraphs (AA), (AB), (AC), (AD), (AE), (AF), (AG), (AH), (AI), (AJ), (AK), (AL), (AM), (AN), (AO) and (AP), further comprising, a time stamp generator operable to generate a time stamp indicating the time at which the digital signature was generated.
  • AR Apparatus in accordance with any of paragraphs (AA), (AB), (AC), (AD), (AE), (AF), (AG), (AH), (AI), (AJ), (AK), (AL), (AM), (AN), (AO), (AP) and (AQ), wherein the sampled data segment is a sampled audio segment.
  • a method of digital signature generation comprising generating an envelope representation, of only one polarity, of a sampled data segment, by
  • AU A method of digital signature generation, comprising generating an envelope representation, of only one polarity, of a sampled data segment, by
  • each threshold value is the sum of (i) the threshold of the preceding portion multiplied by a multiplication factor less than unity, and (ii) the product of the current portion sum value and unity minus said multiplication factor.
  • AX A method in accordance with paragraph (AW), wherein the binary signature is provided by the numerical value or the pattern represented by the values of respective bits of the bit sequence of the binary signature, when those values are taken in a predetermined order as a sequence of binary digits, from most significant to least significant or vice versa.
  • the signature generation comprises, in response to the or a detected event, determining energy content across a frequency spectrum of a section of the sampled data segment, dividing the frequency spectrum into a plurality of frequency bands which are allocated respective identifying numbers, and determining the identifying number of the frequency band of maximum energy, to provide a frequency signature comprising this identifying number.
  • BB A method in accordance with either of paragraphs (AZ) and (BA), comprising band-pass filtering such that the frequency spectrum or spectrums is or are that or those of a band-pass filtered version of the section of the sampled data segment.
  • BC A method in accordance with any of paragraphs (AZ), (BA) and (BB), which is also in accordance with any of paragraphs (AW), (AX) and (AY), wherein the signature generator is operable to provide a signature comprising in combination at least parts of the binary signature and the frequency signature.
  • BD A method in accordance with paragraph (AS) or (AT), or which accordance with any of paragraphs (AW), (AX), (AY), (AZ), (BA), (BB) and (BC) and also in accordance with any of paragraphs (AS) and (AT), further comprising multiplying the par value by a predetermined multiplication factor, to provide the block threshold value.
  • BF A method in accordance with paragraph (BE), comprising band-pass filtering in a plurality of pass bands and generating respective envelope representations on the basis of the data segment as filtered in each pass band.
  • BG A method in accordance with any of paragraphs (AS), (AT), (AU), (AV), (AW), (AX), (AY), (AZ), (BA), (BB), (BC), (BD), (BE) and (BF), comprising detecting, as an event, for each envelope representation generated on the basis of the data segment as filtered in each pass band, a transition of a portion sum value across the threshold value for the portion concerned.
  • BL A computer program operable in computer equipment to cause the equipment to carry out the method in accordance with any of paragraphs (AS), (AT), (AU), (AV), (AW), (AX), (AY), (AZ), (BA), (BB), (BC), (BD), (BE), (BF), (BG), (BH), (BI) and (BJ).
  • AS AS
  • AT AT
  • AU AU
  • AV AV
  • AX AX
  • AY AX
  • AY AX
  • AZ AZ
  • BA BA
  • BB BC
  • BC BC
  • BE BF
  • BG BG
  • BH BH
  • BJ BJ
  • BM A storage medium storing a computer program according to claim (BL).
  • a system comprising a database of digital signatures generated by apparatus or method, as the case may be, in accordance with any of paragraphs (AA) to (BJ), and means for comparing a further digital signature, also generated by apparatus or method, as the case may be, in accordance with any of paragraphs (AA) to (BJ), with digital signatures of the database, to seek a match between said further digital signature and a digital signature of the database.
  • BP A system in accordance with paragraph (BO), wherein said further digital signature is considered to match a compared digital signature of the database if the signatures concerned are bit-wise identical or differ only at selected bit positions of the signatures.
  • BQ A system in accordance with paragraph (BN), wherein permutations of said further digital signature are generated, differing at up to a predetermined number of bit positions from said further digital signature, and a match is considered to be found if said further digital signature or any of the generated permutations are bitwise identical with a digital signature of the database.
  • BR A system in accordance with paragraph (BQ), wherein the permutations differ at up to a predetermined number of predetermined bit positions from said further digital signature.
  • An audience measurement system including, in one or more households, household entertainment equipment such as a TV receiver, radio receiver, or other program source, capable of producing audio signals, and apparatus in accordance with any of paragraphs (AA) to (AR), the apparatus generating digital audio signatures of sampled audio segments derived from the audio signals.
  • reference equipment such as a reference receiver for receiving a plurality of TV and/or radio programs, or other reference program sources comprising audio
  • apparatus in accordance with any of paragraphs (AA) to (AR) the apparatus generating reference digital audio signatures of sampled audio segments derived from the audio of the received programs or other program sources.

Abstract

La présente invention concerne un appareil de production de signatures numériques, comprenant un générateur d'enveloppe apte à produire une représentation d'enveloppe, d'une seule polarité, d'un segment de données échantillonné, et apte à fournir, pour chacune des parties successives comprenant une pluralité prédéterminée d'échantillons, une valeur de somme de partie correspondant à la somme des valeurs des échantillons de la partie, afin de fournir ladite représentation d'enveloppe ; un générateur de valeur de seuil apte à déterminer une valeur de seuil pour chaque partie de la représentation d'enveloppe ; un détecteur d'événement apte à détecter, en tant qu'événement, un passage de la valeur de somme de partie par la valeur de seuil pour la partie concernée ; et un générateur de signature apte à produire, suite à la détection d'un événement, une signature numérique caractéristique du segment de données échantillonné.
EP10750290A 2009-03-11 2010-03-11 Signatures numériques Withdrawn EP2406903A4 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US15912609P 2009-03-11 2009-03-11
PCT/CA2010/000355 WO2010102404A1 (fr) 2009-03-11 2010-03-11 Signatures numériques

Publications (2)

Publication Number Publication Date
EP2406903A1 true EP2406903A1 (fr) 2012-01-18
EP2406903A4 EP2406903A4 (fr) 2013-01-16

Family

ID=42727754

Family Applications (1)

Application Number Title Priority Date Filing Date
EP10750290A Withdrawn EP2406903A4 (fr) 2009-03-11 2010-03-11 Signatures numériques

Country Status (4)

Country Link
US (1) US8769294B2 (fr)
EP (1) EP2406903A4 (fr)
CA (1) CA2754170A1 (fr)
WO (1) WO2010102404A1 (fr)

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8108887B2 (en) 2008-10-30 2012-01-31 The Nielsen Company (Us), Llc Methods and apparatus for identifying media content using temporal signal characteristics
US20140041054A1 (en) * 2012-08-01 2014-02-06 Microsoft Corporation Attestation of possession of media content items using fingerprints
US20140105447A1 (en) * 2012-10-15 2014-04-17 Juked, Inc. Efficient data fingerprinting
US8805865B2 (en) * 2012-10-15 2014-08-12 Juked, Inc. Efficient matching of data
US9183849B2 (en) 2012-12-21 2015-11-10 The Nielsen Company (Us), Llc Audio matching with semantic audio recognition and report generation
US9158760B2 (en) 2012-12-21 2015-10-13 The Nielsen Company (Us), Llc Audio decoding with supplemental semantic audio recognition and report generation
US9195649B2 (en) 2012-12-21 2015-11-24 The Nielsen Company (Us), Llc Audio processing techniques for semantic audio recognition and report generation
CN103763586B (zh) * 2014-01-16 2017-05-10 北京酷云互动科技有限公司 电视节目互动方法、装置和服务器
US9984154B2 (en) * 2015-05-01 2018-05-29 Morpho Detection, Llc Systems and methods for analyzing time series data based on event transitions
US10832692B1 (en) * 2018-07-30 2020-11-10 Amazon Technologies, Inc. Machine learning system for matching groups of related media files

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0210609A2 (fr) * 1985-07-29 1987-02-04 A.C. Nielsen Company Méthode et système d'identification de programmes radiodiffusés
GB2375907A (en) * 2001-05-14 2002-11-27 British Broadcasting Corp An automated recognition system

Family Cites Families (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5436653A (en) * 1992-04-30 1995-07-25 The Arbitron Company Method and system for recognition of broadcast segments
US7095874B2 (en) * 1996-07-02 2006-08-22 Wistaria Trading, Inc. Optimization methods for the insertion, protection, and detection of digital watermarks in digitized data
US7346472B1 (en) * 2000-09-07 2008-03-18 Blue Spike, Inc. Method and device for monitoring and analyzing signals
US7664264B2 (en) * 1999-03-24 2010-02-16 Blue Spike, Inc. Utilizing data reduction in steganographic and cryptographic systems
CA2310769C (fr) * 1999-10-27 2013-05-28 Nielsen Media Research, Inc. Extraction et correlation de signature audio
US20050044561A1 (en) * 2003-08-20 2005-02-24 Gotuit Audio, Inc. Methods and apparatus for identifying program segments by detecting duplicate signal patterns
MX2007002071A (es) * 2004-08-18 2007-04-24 Nielsen Media Res Inc Metodos y aparatos para generar firmas.
US7516074B2 (en) * 2005-09-01 2009-04-07 Auditude, Inc. Extraction and matching of characteristic fingerprints from audio signals
ATE515844T1 (de) * 2006-02-22 2011-07-15 Media Evolution Technologies Inc Verfahren und gerät zum erzeugen von digitalen audiosignaturen
EP2156583B1 (fr) * 2007-05-02 2018-06-06 The Nielsen Company (US), LLC Procédés et appareil de génération de signature
EP2263335B1 (fr) * 2008-03-05 2014-07-23 The Nielsen Company (US), LLC Procédés et appareils de génération de signatures
GB2458471A (en) * 2008-03-17 2009-09-23 Taylor Nelson Sofres Plc A signature generating device for an audio signal and associated methods
WO2009140824A1 (fr) * 2008-05-22 2009-11-26 Yuvad Technologies Co., Ltd. Système conçu pour identifier un contenu vidéo/audio animé
WO2009143667A1 (fr) * 2008-05-26 2009-12-03 Yuvad Technologies Co., Ltd. Système de surveillance automatique des activités de visualisation de signaux de télévision
US8108887B2 (en) * 2008-10-30 2012-01-31 The Nielsen Company (Us), Llc Methods and apparatus for identifying media content using temporal signal characteristics
US8904184B2 (en) * 2008-12-10 2014-12-02 At&T Intellectual Property I, Lp Certification of authenticity of media signals

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0210609A2 (fr) * 1985-07-29 1987-02-04 A.C. Nielsen Company Méthode et système d'identification de programmes radiodiffusés
GB2375907A (en) * 2001-05-14 2002-11-27 British Broadcasting Corp An automated recognition system

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of WO2010102404A1 *

Also Published As

Publication number Publication date
WO2010102404A1 (fr) 2010-09-16
US8769294B2 (en) 2014-07-01
CA2754170A1 (fr) 2010-09-16
EP2406903A4 (fr) 2013-01-16
US20120002806A1 (en) 2012-01-05

Similar Documents

Publication Publication Date Title
US8769294B2 (en) Digital signatures
EP1826932B1 (fr) Méthode et appareil pour la génération de signatures numériques de signaux audiophoniques
US7085613B2 (en) System for monitoring audio content in a video broadcast
US6574594B2 (en) System for monitoring broadcast audio content
EP2263335B1 (fr) Procédés et appareils de génération de signatures
US9136965B2 (en) Methods and apparatus for generating signatures
US6604072B2 (en) Feature-based audio content identification
US7672843B2 (en) Audio signature extraction and correlation
US7031921B2 (en) System for monitoring audio content available over a network
US20110026763A1 (en) Audio visual signature, method of deriving a signature, and method of comparing audio-visual data
EP1550297A1 (fr) Extraction de condenses
EP1567965A1 (fr) Extraction d'empreintes spectrales de contenus multimedia
EP1485907A2 (fr) Systeme de vente d'un produit utilisant une identification de contenu audio
US20210249023A1 (en) Methods and apparatus for audio signature generation and matching
EP1497935A1 (fr) Identification de contenu audio basee sur des caracteristiques
AU2012211498B2 (en) Methods and apparatus for characterizing media
ARMONK Pitman et al.(43) Pub. Date: Jul. 25, 2002

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

17P Request for examination filed

Effective date: 20111011

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO SE SI SK SM TR

DAX Request for extension of the european patent (deleted)
A4 Supplementary search report drawn up and despatched

Effective date: 20121218

RIC1 Information provided on ipc code assigned before grant

Ipc: H04H 60/46 20080101ALI20121212BHEP

Ipc: H04N 7/025 20060101ALI20121212BHEP

Ipc: H04H 60/35 20080101AFI20121212BHEP

Ipc: H04H 60/58 20080101ALI20121212BHEP

GRAP Despatch of communication of intention to grant a patent

Free format text: ORIGINAL CODE: EPIDOSNIGR1

INTG Intention to grant announced

Effective date: 20160422

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE APPLICATION IS DEEMED TO BE WITHDRAWN

18D Application deemed to be withdrawn

Effective date: 20160903